Welcome! to the personal research and interests page of

Daniel P. Ames, PhD, PE


Tulum, Mexico, 2002 Contact Information

E-mail:
Telephone: 208-282-7851
Mail: Idaho State University, 1784 Science Center Dr., Idaho Falls, Idaho, 83402

Current CV/Resume

Resume Overview

I completed my Ph.D. at Utah State University in June 2002, titled "Bayesian Decision Networks for Watershed Management" which examines the use of Bayesian network models of physical and non-physical variables in watershed managment.  My Master's Thesis was completed at USU in June 1998, titled "Seasonal to Interannual Streamflow Forecasts Using Nonlinear Time Series Methods and Climate Information" and focused on the use of nonparametric time series data analysis techniques for long-lead prediction of streamflow with application to the Yakima River Basin in Washington State.  I worked for two years as a Research Assistant Professor at the Utah Water Research Lab (UWRL) at Utah State University in Logan, Utah where helped start and co-direct the Environmental Management Research Group (EMRG). At the EMRG, we developed software and provided training on different environmental management technologies including EPA's BASINS system, and our oringinal geographic information system (GIS) software, MapWindow GIS .  I have also written software components for other programmers and distributed them through my legacy web site Stonehaven Software, and have done programming work for PAQ Services and Worker Rehabilitation Associates, Inc.  In Fall 2004, I took a position at Idaho State University in the Department of Geosciences at the Idaho Falls, Idaho University Place extension campus. I am currently the program coordinator for the Masters of Science in Geographic Information Science at ISU, and I teach classes in GIS, GPS, and Watershed Analysis. My research program continues to develop along the lines of open source GIS tools, watershed modeling and probabilistic methods. Some assorted research related links are provided below.

Probabilistic Modeling and Decision Analysis

Bayesian Decision Networks (BDNs) can be particularly useful to model environmental management problems because often one must bring together a variety of disparate data sources to estimate the relationships between variables in the system. Also, the availability of "hard" data is often limited for generating these estimates, resulting in a case where the model must be estimated from expert elicitation. Thus Bayesian networks are often referred to as "Bayesian Belief Networks" and are populated not with frequency distributions from observed data, but with the best judgement of individuals. Excellent sources of information on Bayesian networks (aside, of course, from my dissertation) include www.hugin.com, www.netica.com and the many references located in my dissertation .

GIS and Decision Support Systems Development

Decision-making is often an interactive, stakeholder-driven process. In light of this, I have worked extensively with environmental data visualization tools linked to watershed, water quality, and other types of models. In particular I am interested in the utility of geographic information systems (GIS) for informing decision-makers and ultimately influencing decision. At the USU we developed a programmable GIS tool, MapWindow GIS , and have used it in several different projects where it continues to be used in projects such as a decision support system for managing the WRIA-1 watershed management area in Washington State. This project integrates many aspects of integrated data analysis and decision-making at multiple levels (stakeholder, scientist, manager, and regulator) and uses technologies and modeling approaches from the purely deterministic to the nonparametric and probabilistic.

7Q10 Low Flow

The 7Q10 low flow statistic is often used as a benchmark whereby to set effluent limits for point source polluters under the NPDES program and point and nonpoint polluters under the TMDL program.  Because of this, the 7Q10 value can have implications for both the environment and industrial and agricultural interests.  An overestimated 7Q10 can have unintended environmental impacts, while an underestimated 7Q10 can have undue economic impacts.  This is a problem that can be addressed in part by computing confidence limits for the 7Q10 estimate.  I have written a paper that introduces a bootstrap approach to estimating 7Q10 confidence limits and compares it to a statistical standard error approach.  I have also written the Low Flow Calculator software for this purpose, an early version of which can be downloaded below.

ISU Courses

Since arriving at ISU in Fall 2004, I have taught and continue to teach a variety of interesting courses (at least I think they are interesting!) Here is the short list:
  • Principles of GIS - an introduction to GIS principles and applications
  • Advanced GIS - more advanced GIS - moving into GIScience
  • GPS Applications in Research - the science and application of GPS
  • Watershed Analysis - application of a variety of geotechnologies to a variety of watershed problems
  • Introduction to GIS Programming - a beginning programming class focused on VB.NET
  • Advanced GIS Programming - ArcObjects, MapWinGIS and more...

Contact Me

You are welcome to contact me by email at:

Thanks for stopping by!

- Dan

 I am nerdier than 94% of all people. Are you a nerd? Click here to find out!

Copyright 2003-2007, Daniel P. Ames